Visualization of Gender Distribution in Philosophy Research Topics


Maximilian Noichl, whose visualizations we’ve discussed previously (here and here) has produced one depicting the gender distribution in philosophy research topics.

The visualization is a rough estimate based on the authorship of over 40,000 papers, using names, to the extent possible, as indicators of gender. It uses color to show the predominant gender in an area (dark green representing a high proportion of men, dark blue representing a high proportion of women, with lighter gradients between). This is what the whole thing, zoomed out, looks like:

The visualization is interactive. To view it go here and click the button that says “show gender distribution.”

If you zoom in on a particular spot, and hover the cursor over it, this is the kind of thing you’ll see:

And then when you click on the highlighted area, you’ll get a list of keywords, journals, and articles in that area:

Overall, the visualization shows what we already knew: that “authorship in philosophy is predominantly male.” But you can search through it to see which areas are more and less gender homogeneous.

Noichl adds: “Please note that it’s still the beta-version, so it’s not as well documented as I want it to be at some point. Feedback welcome.”


Related: posts on philosophy visualizations


guest
18 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
real_email_fake_name
real_email_fake_name
2 years ago

Looks like we have a nice (data-driven and visualized) answer to a question: “[What are] Philosophical Topics of Interest to Women?”. Excellent.

But then again, this was a question for which Justin took a beating for raising some years ago: http://dailynous.com/2014/08/25/philosophical-topics-of-interest-to-women/Report

Maximilian Noichl
Maximilian Noichl
Reply to  real_email_fake_name
2 years ago

I don’t think that my graphic tells us anything about the interests of women or men. It only tells us what the gender-ratios of published authors in which parts of the actual academic landscape actually are. This might have something to do with the interests of women, but it will also have a lot to do with academic culture, with the (dis-)interests of men, with region-specific traditions, etc. Report

Maximilian Noichl
Maximilian Noichl
2 years ago

I don’t think that my graphic tells us anything about the interests of women or men. It only tells us the ratios of published authors in relation to groups of scholarly co-engagement (which tend to be thematic, but are also institutional and methodological). Interests of women obviously will play into that, but so will interests of men (as we’re dealing with ratios), academic culture (e.g. the frequency of multiple authorship), and generally how welcoming different subfields have historically been to women. Report

James
James
2 years ago

Is there a philosophical subfield today that isn’t absolutely desperate to attract the interest of more female scholars? Academics in philosophy departments all over the world are begging on their knees to get women involved in their specialities. It seems like interests of women are a better bet than unwelcome or academic culture, unless the term “academic culture” denotes the apparent lack of interest held by women in many of philosophy’s subfields…Report

Greg conchelos
Greg conchelos
Reply to  James
2 years ago

Interesting to bring in m and fm brain differences although inferences re this are a stretch.

Report

sahpa
sahpa
Reply to  Greg conchelos
2 years ago

James nowhere mentions “brain differences”. Interesting to attribute that to him/her although inferring that is a stretch.Report

Under-represented
Under-represented
Reply to  James
2 years ago

I am a woman that is in of one of those subfields that “women are not interested in”: philosophy of physics and computational sciences. I don’t know how other people in philosophy settled on a subfield, but I imagine that my experience isn’t too dissimilar from most. My choice of subfield resulted from a long series of contingent events. In fact, I went into grad school thinking I would do ethics. I never imagined that I would be doing the type of philosophy of science I do now. The major turning point included finding mentors in philosophy of science that saw my potential and were great teachers. This has a lot to do with academic culture and feeling like you belong. Just having departments “begging” to hire women, doesn’t address any of the real issues. At that point it is too late, people already wrote a dissertation. It is too simplistic to think people choose a subfield solely based off of the *feeling* of interest. I am sure there are other subfields of philosophy that you find interesting, maybe even more interesting than your own, but you never worked on it. Why haven’t you worked on it? Probably because of a long series of contingent events, some of which included you finding a place in philosophy where you felt like you belonged. Report

Avalonian
2 years ago

Thanks for this, lots to note here, including the fact that we’re scheduled to hit parity by 2037 or so. Interesting that the general takeaway appears to be that “authorship is predominantly male” and not that “gendered authorship in philosophy is changing *very* rapidly”. The latter is surely just as striking and deserving of attention, but I suspect we’ve been conditioned to bemoan the state of the discipline and not celebrate its progress.

Max, forgive me if I’m misunderstanding something, but would it be too much work to create a version of this where the percentages are somehow relativized to the actual population numbers? So if 25% of authors are women, treat 25% as being a “middle” number coded white/light-green. The reason I am wondering this is that the image, as it stands, can give the impression that women are generally interested in about three subfields. But of course this is not true: women may be just as likely to gravitate towards a field but that field will show up as green because 75% of authors in general are men right now. The distribution of philosophical interest might be better represented by a relativized chart.Report

Maximilian Noichl
Maximilian Noichl
Reply to  Avalonian
2 years ago

Thank you for your interest. I’m not sure if I share your prediction: There are little black dots in the graph (maybe should’ve been larger) which represent the actual values (opposed to the green line, which are smoothed means). If you look at them, there is indeed a pretty rapid change in the early 2000s — and then they seem to stagnate more or less. Note that the data is unevenly distributed along the time-scale: More papers are sampled from the later period then from the earlier years, which is why the dots are more “bouncy” in the eighties.
I think a lot of the bemoaning comes from the comparision with other disciplines, where philosophy tends to come out regularily as one of the more male dominated disciplines (see e.g. Leslie et al. 2015 for a comparative look into Phd-distributions: https://science.sciencemag.org/content/347/6219/262
Not sure if I agree with their explanation though.)
Nonetheless I agree that the graph does show that change is possible and some has already occured, so that is certainly a valid take-away message.

I’m not sure if I understand your other proposal: Would simply adding the mean (0.47) to every value accomplish what you are thinking about? Then everything that is now beige would be blue — as fields with gender-parity have an overproportional share of woman in relation to the whole — and everything that is now light-green would be beige.
I don’t think the graph can or should be interpreted as representing the *interests* of women, for the reasons I stated above and the reasons the commenter Under-represented has explained — and I don’t think that I indicate anywhere that it should be interpreted this way (although I understand that people are very interested in this). So I feel like the changes you propose (if I understood them correctly) try to get something out of the data that can not be in there.Report

James
James
2 years ago

Under-represented, thanks for your comment and it’s wonderful to hear that things worked out the way they did. That said, I think it’s presumable that there are at least two variables working to determine placement within philosophy: you chose philosophy of science in part because you had great mentors, but also because there was something about the topic that struck you as extremely important and worthy of further inquiry. Then we have to ask whether the gender differential in those fields owes itself to lack of mentorship, varying interest, some combination of the two, or something else entirely. My contention was that lack of mentorship does not seem like a probable culprit, and indeed your experience, as well as the emphasis being placed on creating a diverse work environment in philosophy at the moment, seem to paint a different picture. But if it’s not lack of mentorship, and not different interets that are driving the trend, what is? Report

jdkbrown
jdkbrown
Reply to  James
2 years ago

“…indeed your experience, as well as the emphasis being placed on creating a diverse work environment in philosophy at the moment, seem to paint a different picture.”

This is only right if (1) Under-represented’s experience is typical, and (2) if work towards diversity in Philosophy is substantially finished. I think we have pretty good evidence–in the form of testimony from our colleagues who are most affected by these issues–that neither is the case.Report

James
James
Reply to  jdkbrown
2 years ago

Neither (1) not (2) is relevant to my claim. I mentioned only that Under-represented’s experience in and of itself supports the idea that there is good mentorship for both sexes in philosophy. Where is the evidence that active discrimination against women is keeping (2) from being accomplished? And if it’s passive discrimination, it would be useful to see documented evidence. Report

NoOneNowhere
Reply to  James
2 years ago

James, have you been to graduate school? I mean this seriously (not as some kind of gatekeeping thing.) You don’t go in with your heart absolutely set on any one topic; you will feel drawn to many different areas. An overriding sense of interest is not going to be what ultimately pushes you into one area rather than another. Finding someone you can work with and who you can productively talk about Philosophy with is just as, if not more important. You need to find a topic that grips you, yes, but many will do so, and you more importantly need to find someone who will support you in working on it. The goal is the PhD – it’s a much more pragmatic process than you might expect. I say this because your approach seems a bit naive (lots of “seems to”s and implicit “would have to be”s going on.)Report

James
James
Reply to  NoOneNowhere
2 years ago

Thanks for the response. i’m not sure your statement is at odds with mine. No doubt mentorship also plays a role in sub-specialization. Another point is that even then it’s still not clear that mentorshop’s playing a large role does anything to minimize the large role of interest (would you disagree? in determining choice of research topics. Regarding your second point: which of the assumptions you feel I’ve made would you be averse to making yourself? Do you have evidence to the contrary? Report

James
James
Reply to  James
2 years ago

Re grad school-I didn’t go. Who knows if I’d have gotten in. Largely because, of the graduate students who were my TAs, not a single one now has a job teaching philosophy. That despite having taught at a prestigious program. Was very hard to watch. Report

James
James
Reply to  James
2 years ago

That should of course be mentorship*. Report

Spencer
Spencer
2 years ago

I think it’s fair to say that this data is defeasible evidence of male/female differences in interest. It’s naturalistic and there are variables that haven’t been isolated as others have pointed out. But the standard for evidence here shouldn’t be irrefutable proof. Maximilian acknowledged “Interests of women obviously will play into that…” Right. Unless we’re saying that things like stereotypes and masculine subcultures within philosophy explain absolutely all of this variation. My priors for that are low.

If this doesn’t even constitute defeasible evidence of differences in interest, then I wonder what would. What kind of data would we have to have before we could legitimately say “Ok, this isn’t proof, but it sure looks like evidence that men and women in philosophy tend to have different interests on average”?
Report

Maximilian Noichl
Maximilian Noichl
Reply to  Spencer
2 years ago

I’m not sure if anyone doubts that one could find some differences in philosophical interest along the gender-axis. I think there would be several appropriate methods to do so, for example surveys at different career stages that ask what the student or scholar in question always would have liked to learn more about, if he or she had had the time/resources/mentoring, etc.

I think these results would have to be taken with the usual qualifications regarding gender differences — they certainly do not tell as anything about male and female brains for example, although some commenter above seems to think so. But they might be interesting, at least if the survey is well done — and I don’t think I could predict the outcome.

The question regarding the visualization is of course, how much of the differences in actually published scholarship are explained by these differences in interest. As there are multiple other variables that will play into this — you mention “stereotypes and masculine subcultures”, and I would, off the cuff, add to that stuff like:
* the age of the subfield (when there were recently a lot of new positions to fill, that might make a big difference)
* frequency of co-authorship
* interdisciplinarity with other sciences (I imagine it makes a difference, whether people cooperate more with psychologists or physicists)
* sub-field-specific institutions

All of theses things probably influence our final publication-outcome. And as we can from my graphic only see that outcome, we can not really tell how much of that variance is explained by gender-specific interests — it might be 8% or it might be 80%. But without knowing that number, we can’t really make inferences about interest from publication outcomes.

Report